12734 Phys. Chem. Chem. Phys., 2013, 15, 12734--12741 This journal is c the Owner Societies 2013

Cite this: Phys. Chem.Chem.Phys.,2013,15, 12734

Specific interactions between the quaternaryammonium oligoether-based ionic liquid andwater as a function of pressure

Hai-Chou Chang,*a Jyh-Chiang Jiang,*b Tsai-Yi Chen,a Hsing-Sheng Wang,a

Leo Yuxiu Li,a Wei-Wen Hunga and Sheng Hsien Linc

The interactions between Ammoeng 100 and water are probed using high-pressure infrared measurements

and DFT-calculations. The results of infrared absorption profiles suggest that the energetically favored

approach for water molecules to interact with Ammoeng 100 is via the formation of anion–water inter-

actions, whereas the alkyl C–H groups play much less important roles. After comparison with pure

Ammoeng 100, it appears that no appreciable changes in band frequencies of alkyl C–H vibrations

occurred as Ammoeng 100 was mixed with D2O. The presence of D2O has a red-shift effect on the peak

frequency of the SQO stretching vibration under the pressures below 1 GPa in comparison to the

absorption frequencies of pure Ammoeng 100. This observation is likely related to local structures of

the SQO groups interacting with D2O molecules. DFT-calculations indicate that the most energetically

favored conformation of ion pairs should be the species having only one hydrophilic hydrogen bonding.

The results of calculations reveal that water addition may induce the partial replacement of C–H� � �Ointeractions with strong hydrogen bonding between anions and water molecules.

Introduction

Ionic liquids are typically composed of bulky organic cationsand anions with a melting temperature below 100 1C. Theirextraordinary properties such as nonvolatile nature and liquidstate in a broad temperature range make them attractivealternatives to volatile organic solvents for many innovativeapplications.1–3 Although ionic liquids are environmentallygreen solvents compared to traditional organic solvents, thehigh viscosity of ionic liquids has restricted their use assolvents in technical applications such as heat-transfer fluids.To expand the application of ionic liquids, addition of asuitable co-solvent to an ionic liquid has come into focus inrecent years.4–13 The study of ionic liquid-based mixturesbecomes more and more attractive in recent years and this isone of the motivations for the current work.1–13

The properties of ionic liquid-based mixtures are very complexthat have been shown using experimental and theoretical methods.

Generally, the structures and properties of these ionic systemsare mainly dominated by the electrostatic interactions.2,5 Basedon previous studies, ionic liquids differ from the classical saltsat least in one aspect: they possess hydrogen bonds that inducestructural directions.2,5 However, the role of hydrogen bondingin ionic liquids is still an issue of debate in the literature.5,14

The local structures and self-assembly of ionic liquid–watermixtures have been studied, while the physical properties arekey features for the applications of ionic liquids. At high ionicliquid concentrations, ionic liquids seem to form clusters, as inthe pure state, and water molecules interact with the clusterswithout interacting among themselves. Ionic liquids containingdissolved water may not be regarded as homogeneous solventsbut have to be considered as microheterogeneous materials,where the microheterogeneity may be attributed to incompletemixing at the molecular level.15–18

Because ionic liquids are entirely composed of ions, ionicliquids have been utilized in the preparation of ion gels (a new classof solid state electrolytes).19–22 The addition of gelling agents to ionicliquids occasionally leads to the gelation of ionic liquids and thegelling agents include low molecular weight gelators, nanoparticles,and block copolymers. Such ion gels, prepared by chemical orphysical gelation, frequently exhibit intriguing properties (e.g., highionic conductivity, quasi-solidification, and flexibility) making

a Department of Chemistry, National Dong Hwa University, Shoufeng, Hualien 974,

Taiwan. E-mail: hcchang@mail.ndhu.edu.twb Department of Chemical Engineering, National Taiwan University of Science and

Technology, Taipei 106, Taiwan. E-mail: jcjiang@mail.ntust.edu.twc Department of Applied Chemistry, National Chiao Tung University,

Hsinchu 30010, Taiwan

Received 3rd April 2013,Accepted 30th May 2013

DOI: 10.1039/c3cp51396c

www.rsc.org/pccp

PCCP

PAPER

Publ

ishe

d on

30

May

201

3. D

ownl

oade

d by

Nat

iona

l Chi

ao T

ung

Uni

vers

ity o

n 28

/04/

2014

01:

57:0

8.

View Article OnlineView Journal | View Issue

This journal is c the Owner Societies 2013 Phys. Chem. Chem. Phys., 2013, 15, 12734--12741 12735

them attractive for the application of ion gels. However, thepreparation of ion gels suffers from the diverse limitationsof expensive synthetic procedures and unknown potentialtoxicity. Developing new gelation methods for ionic liquids iscrucial for the significant application of ion gels. There hasrecently been much interest in using water to prepare iongels and this inexpensive approach yields ion gels with goodphysical and mechanical properties.6,7 Recently the gelationof the quaternary ammonium oligoether-based ionic liquidAmmoeng 100 with water has been addressed.6,7 Nevertheless,the local structures of water-based ion gels are not clearat present. In this work, we use variable pressure as a windowinto the hydrogen bonding structures in Ammoeng 100–watermixtures.

Knowledge of the nature of hydrogen bonding interactionsis fundamental to understand the physical properties of ionicliquids. The hydrogen bonding interactions are complex forionic liquid–water mixtures with varying anions, due to pro-nounced anion–water interactions. The molecular state ofwater absorbed from air in ionic liquids has been studied usingvibrational spectroscopy, which indicates that water moleculesinteract with the anions and exist in symmetric 1 : 2 typehydrogen-bonded complexes.23,24 Many research studies ofionic liquid mixtures are interested in the interactions betweenwater and anions, while the interactions of water with cationshave been studied limitedly. Water miscibility of ionic liquids isaffected by the length of alkyl side-chains and the role of waterin ionic liquids is closely related to the supramolecular struc-tures of ionic liquids. Researchers have demonstrated thatsmall amounts of water in ionic liquids have a dramatic effecton the rate of diffusion.17 Based on the theoretical results ofVoth’s group,15 at high ionic liquid concentration ionic liquidsseem to form clusters, as in the pure state, and water moleculesinteract with the clusters without interacting among them-selves. In the present investigation, our intent is to achievefurther understanding of aggregation properties in ionic liquid–water mixtures.

Various spectroscopic and theoretical studies have been madeto probe the structures, interactions, and solvation of ionicliquid-based systems.1–5 Previous studies on the structures ofionic liquids have included the use of X-ray crystallography.1

Although the results of crystal structures are highly informativeon the relative geometry changes, crystallography does notprovide direct information on the local structures in the liquidstate. Thus, vibrational spectroscopy (IR or Raman) is oftenused to explore hydrogen bonded structures of liquid mixtures.Generally, vibrational studies were performed at ambient pres-sure and mostly at room temperature. Recently interest inpressure as an experimental variable has been growing inphysicochemical studies.21,22,25–27 Under high-pressure condi-tions, the relative weights of the strong intramolecular inter-actions responsible for molecular bonding and of the weakerintermolecular forces defining the aggregation states arealtered. In this study, we show that high-pressure infraredspectroscopy is a sensitive method to probe the structuralorganization in ionic liquid–water mixtures.

Experimental section

Samples were prepared using Ammoeng 100 (>95%, UniRegionBio-Tech) and D2O (99.9 atom% D, Aldrich). A diamond anvilcell (DAC) of Merrill–Bassett design, having a diamond culet sizeof 0.6 mm, was used for generating pressures up to ca. 2 GPa.Two type-IIa diamonds were used for mid-infrared measure-ments. The sample was contained in a 0.3 mm-diameter holein a 0.25 mm-thick inconel gasket mounted on the diamondanvil cell. To reduce the absorbance of the samples, CaF2 crystals(prepared from a CaF2 optical window) were placed into the holeand compressed firmly prior to inserting the sample. A droplet ofa sample filled the empty space of the entire hole of the gasket inthe DAC, which was subsequently sealed when the opposedanvils were pushed toward one another. Infrared spectra of thesamples were measured on a Perkin-Elmer Fourier transformspectrophotometer (model Spectrum RXI) equipped with a LITA(lithium tantalite) mid-infrared detector. The infrared beam wascondensed through a 5� beam condenser onto the sample in thediamond anvil cell. Typically, we chose a resolution of 4 cm�1

(data point resolution of 2 cm�1). For each spectrum, typically1000 scans were compiled. To remove the absorption of thediamond anvils, the absorption spectra of DAC were measuredfirst and subtracted from those of the samples. Pressure calibra-tion follows Wong’s method.28,29 The pressure dependence ofscrew on moving distances was measured.

Results and discussion

The quaternary ammonium-based ionic liquid chosen for thepreparation of the mixture in this article is Ammoeng 100, asshown in Fig. 1. In general, Ammoeng ionic liquids haveamphiphilic structures containing both long alkyl chainsand hydroxyl groups. Fig. 2 displays infrared spectra of pureAmmoeng 100 obtained under ambient pressure (curve a) and

Fig. 1 Chemical structures of (a) Ammoeng 100 and (b) the cation used in theDFT-calculation.

Paper PCCP

Publ

ishe

d on

30

May

201

3. D

ownl

oade

d by

Nat

iona

l Chi

ao T

ung

Uni

vers

ity o

n 28

/04/

2014

01:

57:0

8.

View Article Online

12736 Phys. Chem. Chem. Phys., 2013, 15, 12734--12741 This journal is c the Owner Societies 2013

at 0.3 (curve b), 0.9 (curve c), 1.5 (curve d), 1.9 (curve e),2.3 (curve f), and 2.5 GPa (curve g). We have concentrated ouranalysis on the 2800–3800 cm�1 region which contains the C–Hand O–H stretching modes, Fig. 2.21,22 The infrared spectrumexhibits two discernible peaks, i.e., at 2857 and 2926 cm�1,respectively, corresponding to C–H stretching modes of the alkylgroups, Fig. 2a. The broad band at ca. 3401 cm�1 in Fig. 2a can beassigned to hydrogen bonded O–H stretching modes.30 As thesample was compressed, i.e., increasing from ambient (Fig. 2a) to0.3 GPa (Fig. 2b), alkyl C–H bands were blue shifted to 2863 and2935 cm�1, respectively, Fig. 2b. The blue shift may originate fromthe combined effect of the overlap repulsion enhanced by hydro-static pressure, C–H� � �O contacts, and so forth.21,22,31–33 Never-theless, the red shift of the O–H band at ca. 3374 cm�1 is obviousas the pressure was elevated, Fig. 2b. This behavior is in accordwith the general trend observed for a red shift with pressure forO–H and CQO stretching modes in strongly hydrogen bondedO–H� � �O and CQO� � �H systems, respectively.30 In contrast tostrong hydrogen bonding, the C–H covalent bond tends to blue-shift as a result of formation of a weak hydrogen bond.

Fig. 3 presents infrared spectra of an Ammoeng 100–D2Omixture (10 wt% Ammoeng 100) obtained under ambient pres-sure (curve a) and at 0.3 (curve b), 0.9 (curve c), 1.5 (curve d),

1.9 (curve e), 2.3 (curve f), and 2.5 GPa (curve g). The C–Hstretching absorptions overlap with the O–H stretching bands ofH2O, so D2O was used in this study. By comparing Fig. 2a and 3a,we observe no appreciable changes in C–H spectral features inthe presence of D2O under ambient pressure, Fig. 3a. The broadabsorption band at ca. 3406 cm�1 revealed in Fig. 3a is mainlyattributed to the hydrogen bonded O–H stretching modes of thediluted HOD in the Ammoeng 100–D2O mixture. The HODmolecules may mainly originate from the H–D exchange fromD2O to the cation. HOD is also a typical impurity existing in D2O.As the pressure was elevated to 0.3 GPa, the red shift of thehydrogen bonded O–H band is obvious, Fig. 3b. At a pressure of1.5 GPa (Fig. 3d), the O–H spectral features show further evolu-tion through an observation of bandwidth narrowing. Thisobservation indicates that a pressure-induced phase transitionoccurs, Fig. 3d. At room temperature, compression of liquidwater leads to tetragonal ice VI at pressure above 1.05 GPa andfurther to cubic ice VII above 2.1 GPa.30 The sharper O–Hfeatures, Fig. 3d–g, may correspond to the high order environ-ment in ice-like structures.30 The IR spectra of both ice VI(Fig. 3d–e) and ice VII (Fig. 3f–g) consist of two O–H bands

Fig. 2 IR spectra of pure Ammoeng 100 obtained under ambient pressure(curve a) and at 0.3 (curve b), 0.9 (curve c), 1.5 (curve d), 1.9 (curve e), 2.3 (curve f),and 2.5 GPa (curve g).

Fig. 3 IR spectra of an Ammoeng 100–D2O mixture (10 wt% Ammoeng 100)obtained under ambient pressure (curve a) and at 0.3 (curve b), 0.9 (curve c),1.5 (curve d), 1.9 (curve e), 2.3 (curve f), and 2.5 GPa (curve g).

PCCP Paper

Publ

ishe

d on

30

May

201

3. D

ownl

oade

d by

Nat

iona

l Chi

ao T

ung

Uni

vers

ity o

n 28

/04/

2014

01:

57:0

8.

View Article Online

This journal is c the Owner Societies 2013 Phys. Chem. Chem. Phys., 2013, 15, 12734--12741 12737

overlapping each other. These two bands become well separatedin the derivative spectra. As shown in Fig. 3b–g, the compressionleads to blue frequency shifts of the C–H stretching modes.

The absorption frequencies of the dominant alkyl C–H bandat about 2926 cm�1 are plotted versus pressures in Fig. 4, whilewe notice that the alkyl C–H band in Fig. 4 displays anomalousnon-monotonic pressure-induced frequency shifts. For pureAmmoeng 100, the compression leads to blue frequency shiftsat the pressure below 0.3 GPa, then no appreciable changes inthe band frequency (cf. Fig. 2b–c), and then blue-shifts againupon further increasing the pressure (cf. Fig. 2d–g). This dis-continuity in frequency shift, Fig. 4, is in agreement with thetrends revealed in our previous studies.34 The alkyl C–H bandfrequency of pure Ammoeng 100 increases significantly at0.3 GPa, Fig. 4, and this may indicate a pressure-inducedsolidification (or structural transformation) at a pressure of0.3 GPa. A structural reorganization or second phase transitionmay take place at pressures above 1 GPa as revealed in Fig. 4.After comparison with pure Ammoeng 100, it appears that noappreciable changes in band frequencies of alkyl C–H vibra-tions occurred as Ammoeng 100 was mixed with D2O, Fig. 4.Our results in Fig. 4 suggest that the presence of D2O does notperturb the ionic liquid–ionic liquid clustering (phase transi-tion or cross-linking) in the non-polar region at the pressure of0.3 GPa. The role of water in ionic liquids is complex anddepends on the supramolecular structures of ionic liquids.15–17

We would like to point out that the HOD molecules in theAmmoeng 100–D2O mixture are still liquid-like under the pressureof 0.3–1 GPa as revealed in Fig. 3b and c. Thus, the Ammoeng100–D2O mixture, Fig. 3b and c may behave like gel due tothe pressure-induced transition via a three-dimensional cross-linked network within the liquid D2O. Our results indicate thathigh pressure can be applied to tune the relative weights ofionic liquid aggregation states in the Ammoeng 100–D2Omixture even in the liquid (or gel) state.

In order to gain further insights into the local structures ofionic liquids, the pressure dependencies of SQO stretchingvibration of pure Ammoeng 100 and Ammoeng 100–D2O arepresented in Fig. 5 and 6, respectively. Fig. 5 displays infrared

spectra of pure Ammoeng 100 obtained under ambient pres-sure (curve a) and at 0.3 (curve b), 0.9 (curve c), 1.5 (curve d),1.9 (curve e), 2.3 (curve f), and 2.5 GPa (curve g) in the980–1080 cm�1 region. As revealed in Fig. 5a, the symmetricSO3 stretch appears at 1013 cm�1 under ambient pressure.35,36

The symmetric SO3 stretch shows a red shift to 1011 cm�1 uponcompression, Fig. 5b and this result represents the weakeningof the SQO bond. The sharper SQO structure revealed inFig. 5b is in part due to the higher order and anisotropicenvironment in a solid structure. As the sample was com-pressed, i.e., increasing the pressure from 0.3 GPa (Fig. 5b) to2.3 GPa (Fig. 5e), no appreciable changes in the band frequencyof the SQO stretching absorption were observed.

Fig. 6 displays infrared spectra of an Ammoeng 100–D2Omixture (10 wt% Ammoeng 100) obtained under ambient pressure(curve a) and at 0.3 (curve b), 0.9 (curve c), 1.5 (curve d), 1.9 (curve e),2.3 (curve f), and 2.5 GPa (curve g) in the 980–1080 cm�1 region.A red shift in band frequency of the SQO stretching vibrationfrom 1013 cm�1 (Fig. 5a) to 1006 cm�1 (Fig. 6a) occurred as theionic liquid was diluted by D2O under ambient pressure. Thisresult is remarkably different from what is revealed for the alkylC–H groups, Fig. 3a. The SQO stretching mode underwent a

Fig. 4 Pressure dependence of the C–H stretching frequencies of pure Ammoeng100 (diamonds) and the Ammoeng 100–D2O mixture (squares).

Fig. 5 IR spectra of pure Ammoeng 100 obtained under ambient pressure(curve a) and at 0.3 (curve b), 0.9 (curve c), 1.5 (curve d), 1.9 (curve e), 2.3 (curve f),and 2.5 GPa (curve g).

Paper PCCP

Publ

ishe

d on

30

May

201

3. D

ownl

oade

d by

Nat

iona

l Chi

ao T

ung

Uni

vers

ity o

n 28

/04/

2014

01:

57:0

8.

View Article Online

12738 Phys. Chem. Chem. Phys., 2013, 15, 12734--12741 This journal is c the Owner Societies 2013

red-shift in frequency as D2O was added, Fig. 6a, but nosignificant changes in C–H stretching frequency were observedin Fig. 3a under ambient pressure. Previous studies have shownthat the frequency shifts of the vibration modes are closelyrelated to changes in the hydration states and the liquidstructures.34 Our results may suggest that the energeticallyfavored approach for the D2O molecules to interact withAmmoeng 100 is via the formation of anion–D2O interactions.The compression leads to a red shift in frequency of the SQOstretching vibration to 1004 cm�1, Fig. 6b, while blue-shifts infrequency of the SQO stretching vibration are observed forfurther compressions, Fig. 6c–g. Looking into more detail inFig. 6d–g, the spectral features of the SQO stretching vibrationof the Ammoeng 100–D2O mixture show further evolution upondilution through the observation of decreases in band widths incomparison to the SQO absorption bandwidths of pureAmmoeng 100 as revealed in Fig. 5d–g. Our results indicatethat the SQO absorption bandwidths in Fig. 5 and 6 are moreconcentration-sensitive than the alkyl C–H bandwidths asshown in Fig. 2 and 3. A possible explanation for this effect is

the hydrogen bonding formation between the SQO groups andD2O. The high order ice-like O–D structures may be one of thereasons to yield the bandwidth narrowing, Fig. 6d–g. Thisobservation suggests the formation of a certain ice-like struc-ture around the SQO groups, but the details remain unclear.Based on our results, the SQO groups seem to be morefavorable sites for hydrogen bonding than the alkyl groups.We also observed several C–O stretching bands in the1020–1080 cm�1 region. One of the C–O peaks locating between1020 and 1030 cm�1 becomes obvious, Fig. 6 (in comparison toFig. 5), due to the bandwidth narrowing of SQO absorption.

To illustrate the frequency shift, the absorption frequenciesof the SQO stretching vibration was presented versus pressuresin Fig. 7 and the SQO stretching band displays anomalous non-monotonic pressure-induced frequency shifts. As shown inFig. 7, the presence of D2O has a red-shift effect on the peakfrequency of the SQO stretching vibration under the pressuresbelow 1 GPa in comparison to the absorption frequencies ofpure Ammoeng 100. This observation is likely related to localstructures of the SQO groups interacting with D2O molecules.We note that simple density effects may result in monotonicfrequency shifts due to increased density upon compression. Inother words, the non-monotonic frequency shifts observed inthis study suggest the changes in structures due to hydrogenbonding. The red-shifts of SQO stretches may be due to thereplacement of an interaction between the anion and a cationwith an interaction with D2O. We note that the SQO absorptionpeaks of the Ammoeng 100–D2O mixture show the trend of aninitial red shift followed by blue shifting under compression,Fig. 7. As pure Ammoeng 100 and the Ammoeng 100–D2Omixture were compressed to the pressures above 1 GPa,Fig. 7, the symmetric SO3 absorption displays almost the sameband frequencies and the presence of D2O does not have adrastic effect on the band frequencies of SQO stretches as theAmmoeng 100 was diluted by D2O. This result is remarkablydifferent from the lower SO3 frequency observed for theAmmoeng 100–D2O mixture under the pressures below 1 GPa,Fig. 7. This behavior may arise from changes in water-cluster

Fig. 6 IR spectra of an Ammoeng 100–D2O mixture (10 wt% Ammoeng 100)obtained under ambient pressure (curve a) and at 0.3 (curve b), 0.9 (curve c),1.5 (curve d), 1.9 (curve e), 2.3 (curve f), and 2.5 GPa (curve g).

Fig. 7 Pressure dependence of the symmetric SO3 stretching frequencies ofpure Ammoeng 100 (diamonds) and the Ammoeng 100–D2O mixture (squares).

PCCP Paper

Publ

ishe

d on

30

May

201

3. D

ownl

oade

d by

Nat

iona

l Chi

ao T

ung

Uni

vers

ity o

n 28

/04/

2014

01:

57:0

8.

View Article Online

This journal is c the Owner Societies 2013 Phys. Chem. Chem. Phys., 2013, 15, 12734--12741 12739

sizes (ice-like structures) and a perturbation in the geometricalproperties of hydrogen-bond network as the pressure iselevated to above 1 GPa. The association of water moleculesto form the ice-like structures may induce the clustering ofAmmoeng 100. In the literature, the researchers have reportedthat the cation–anion interaction can be distinguished fromanion–water or cation–water interactions using vibrationalspectroscopy,37 and ionic liquid–water mixtures are not anunstructured homogeneous solution.38

To develop further insight into the hydration mechanism,we have performed density functional theory (DFT) calculationsusing the Gaussian program package.39 As compared with thechemical structure of Ammoeng 100, Fig. 1a, we choose theshorter side-chain cation, having two methyl groups and twoside-chains with m = n = 1, in our calculations (see Fig. 1b).Fig. 8 displays the DFT-calculated structures of [cation][anion]ion pairs (species A, B, and C), ([cation][anion])2 ion-pair dimer(species C-d), ([cation][anion])2(water) (species C-d-w), and([cation][anion])2(water)2 (species C-d-w2). The geometry opti-mization was calculated by using the B3LYP level with 6-31+G*.Table 1 displays energy results and predicted symmetric SO3

stretching frequencies. As illustrated in Fig. 8A and B, the approach

for the anion (CH3SO4�) to interact with the cation is through the

formation of two hydrophilic hydrogen bonds (–O–H� � �OQS–)for species A and hydrophobic interactions (–C–H� � �OQS–) forspecies B. Based on calculated results revealed in Table 1, thehydrophilic conformer (species A) is more stable in thermo-dynamic energy than the hydrophobic conformer (species B).Nevertheless, Table 1 indicates that the most energeticallyfavored conformation of ion pairs should be species C havingonly one hydrophilic hydrogen bonding (–O–H� � �OQS–). Thisfact suggests the non-negligible role of weak hydrogen bondssuch as C–H� � �O interactions in the structures of ion pairs.

Fig. 8 Optimized structures of ion pairs (species A, B, and C), the ion-pair dimer (species C-d), ion-pair-(water) (species C-d-w), and ion-pair-(water)2 (species C-d-w2).

Table 1 DFT-calculated relative energies (hartree mol�1) and symmetric SO3

stretching frequencies (cm�1)a

Speciesa Relative energies Calcd frequency

A �1489.637914 979.9B �1489.636111 985.0C �1489.645719 992.0C-d �2979.307899 988.4, 992.0C-d-w �3055.722625 985.4, 991.8C-d-w2 �3132.137958 985.0, 986.8

a Structures illustrated in Fig. 8.

Paper PCCP

Publ

ishe

d on

30

May

201

3. D

ownl

oade

d by

Nat

iona

l Chi

ao T

ung

Uni

vers

ity o

n 28

/04/

2014

01:

57:0

8.

View Article Online

12740 Phys. Chem. Chem. Phys., 2013, 15, 12734--12741 This journal is c the Owner Societies 2013

It appears that charge-enhanced alkyl C–H� � �O interactionssomehow provide complementary stabilization energies to pro-vide stability to the local structures of neat ionic liquids.

As the clusters increase in size, the number of low-lyingisomers increases exponentially and the structure identifi-cation is complicated. Thus, we only extend the structure ofspecies C to develop the structure of the ion-pair dimer (speciesC-d), dimer-water (species C-d-w), and dimer-(water)2 (speciesC-d-w2), Fig. 8. As revealed in the structures of species C-d-wand C-d-w2, water addition disrupts C–H� � �O interactionsbetween the cation and the anion. It was known that theaddition of an appropriate amount of water to Ammoeng 100results in the formation of ionogel, but the detailed mechanismis still not clear. Based on our calculations of species C-d-w andC-d-w2, the gelation of Ammoeng 100/water may involvethe partial replacement of C–H� � �O interactions with stronghydrogen bonding between the anion and water molecules.Table 1 displays the frequencies of two symmetric SO3 stretch-ing vibrations located at 988.4 and 992.0 cm�1 for species C-dand the symmetric SO3 stretching bands are red-shifted to985.4 and 991.8 cm�1 for species C-d-w. As species C-d iscoordinated to two water molecules, the symmetric SO3 stretchingbands are further red-shifted to 985.0 and 986.8 cm�1 for speciesC-d-w2. This result is in agreement with the red-shift observedupon dilution under ambient pressure, Fig. 7.

Conclusion

The pressure-dependent infrared spectral features indicate thatthe alkyl C–H bands of Ammoeng 100 display non-monotonicpressure-induced frequency shifts. The compression leads toblue frequency shifts at the pressure below 0.3 GPa, then noappreciable changes in the band frequency, and then blue-shifts again upon further increasing the pressures. No signifi-cant changes in alkyl C–H stretching frequencies were observedas Ammoeng 100 was mixed with D2O. This observation sug-gests that the presence of D2O does not perturb the ionicliquid–ionic liquid association in the non-polar region. Incontrast to the C–H groups, the SQO stretching modes under-went red-shift in frequency in the presence of D2O under thepressures below 1 GPa. Our results suggest that the energeti-cally favored approach for D2O molecules to interact withAmmoeng 100 is via anion–D2O interactions. DFT-calculationsindicate the non-negligible role of hydrophobic interactions(C–H� � �O) in the structures of ion pairs. Water additiondisrupts C–H� � �O interactions and the partial replacementof hydrophobic interaction with strong hydrogen bondingbetween anion and water molecules occurs in Ammoeng100–water mixtures.

Acknowledgements

The authors thank the National Dong Hwa University and theNational Science Council (Contract No. NSC 101-2113-M-259-006-MY3) of Taiwan for financial support. The authors thankYu-Sheng Cheng and Rong-Lin Zhang for the assistance.

Notes and references

1 Ionic Liquids in Synthesis, ed. P. Wasserscheid andT. Welton, Wiley VCH, Weinheim, Germany, 2008.

2 J. Dupont and J. D. Scholten, Chem. Soc. Rev., 2010, 39,1780.

3 E. W. Castner and J. F. Wishart, J. Chem. Phys., 2010, 132,120901.

4 M. Brehm, H. Weber, A. S. Pensado, A. Stark andB. Kirchner, Phys. Chem. Chem. Phys., 2012, 14, 5030.

5 A. Wulf, K. Fumino and R. Ludwig, Angew. Chem., Int. Ed.,2010, 49, 449.

6 J. C. Ribot, C. Guerrero-Sanchez, R. Hoogenboom andU. S. Schubert, J. Mater. Chem., 2010, 20, 8279.

7 P. Reddy, K. J. Chiyen, N. Deenadayalu and D. Ramjugernath,J. Chem. Thermodyn., 2011, 43, 1178.

8 S. Sarkar, R. Pramanik, C. Ghatak, P. Setua and N. Sarkar,J. Phys. Chem. B, 2010, 114, 2779.

9 Q. G. Zhang, N. N. Wang and Z. W. Yu, J. Phys. Chem. B,2010, 114, 4747.

10 K. S. Rao, T. Singh, T. J. Trivedi and A. Kumar, J. Phys. Chem.B, 2011, 115, 13847.

11 P. K. Nayak, A. P. Hathorne and H. Bermudez, Phys. Chem.Chem. Phys., 2013, 15, 1806.

12 Y. Fukaya and H. Ohno, Phys. Chem. Chem. Phys., 2013,15, 4066.

13 W. Shi, K. Damodaran, H. B. Nulwala and D. R. Luebke,Phys. Chem. Chem. Phys., 2012, 14, 15897.

14 K. Fumino, T. Peppel, M. Geppert-Rybczynska, D. H.Zaitsau, J. K. Lehmann, S. P. Verevkin, M. Kockerling andR. Ludwig, Phys. Chem. Chem. Phys., 2011, 13, 14064.

15 Y. Wang and G. A. Voth, J. Phys. Chem. B, 2006, 110, 18601.16 J. N. A. C. Lopes and A. A. H. Padua, J. Phys. Chem. B, 2006,

110, 3330.17 U. Schroder, J. D. Wadhawan, R. G. Compton, F. Marken,

P. A. Z. Suarez, C. S. Consorti, R. F. de Souza and J. Dupont,New J. Chem., 2000, 24, 1009.

18 M. Tanaka, T. Yago and M. Wakasa, Phys. Chem. Chem.Phys., 2013, 15, 787.

19 H. Park, H. S. Kim and Y. M. Jung, J. Phys. Chem. B, 2011,115, 1743.

20 S. Taguchi, T. Matsumoto, T. Ichikawa, T. Kato andH. Ohno, Chem. Commun., 2011, 47, 11342.

21 H. C. Chang, T. C. Hung, S. C. Chang, J. C. Jiang andS. H. Lin, J. Phys. Chem. C, 2011, 115, 11962.

22 H. C. Chang, S. C. Chang, T. C. Hung, J. C. Jiang, J. L. Kuoand S. H. Lin, J. Phys. Chem. C, 2011, 115, 23778.

23 M. Lopez-Pastor, M. J. Ayora-Canada, M. Valcarcel andB. Lendl, J. Phys. Chem. B, 2006, 110, 10896.

24 L. Cammarata, S. G. Kazarian, P. A. Salter and T. Welton,Phys. Chem. Chem. Phys., 2001, 3, 5192.

25 J. C. Jiang, K. H. Lin, S. C. Li, P. M. Shih, K. C. Hung,S. H. Lin and H. C. Chang, J. Chem. Phys., 2011, 134,044506.

26 H. C. Chang, T. C. Hung, H. S. Wang and T. Y. Chen,AIP Adv., 2013, 3, 032147.

PCCP Paper

Publ

ishe

d on

30

May

201

3. D

ownl

oade

d by

Nat

iona

l Chi

ao T

ung

Uni

vers

ity o

n 28

/04/

2014

01:

57:0

8.

View Article Online

This journal is c the Owner Societies 2013 Phys. Chem. Chem. Phys., 2013, 15, 12734--12741 12741

27 Y. Umebayashi, J. C. Jiang, K. H. Lin, Y. L. Shan, K. Fujii,S. Seki, S. Ishiguro, S. H. Lin and H. C. Chang, J. Chem.Phys., 2009, 131, 234502.

28 P. T. T. Wong, D. J. Moffatt and F. L. Baudais, Appl.Spectrosc., 1985, 39, 733.

29 P. T. T. Wong and D. J. Moffatt, Appl. Spectrosc., 1987, 41, 1070.30 H. C. Chang, K. H. Huang, Y. L. Yeh and S. H. Lin, Chem.

Phys. Lett., 2000, 326, 93.31 D. Liu, W. Lei, K. Wang, G. Bao, F. Li, J. Hao, B. Liu, T. Cui,

Q. Cui and G. Zou, J. Phys. Chem. B, 2009, 113, 7430.32 M. Jablonski and A. Sadlej, Chem. Phys. Lett., 2008, 463, 322.33 O. Russina, B. Fazio, C. Schmidt and A. Triolo, Phys. Chem.

Chem. Phys., 2011, 13, 12067.34 H. C. Chang, J. C. Jiang, W. C. Tsai, G. C. Chen and S. H. Lin,

J. Phys. Chem. B, 2006, 110, 3302.35 Y. Miller, G. M. Chaban, J. Zhou, K. R. Asmis, D. M. Neumark

and R. B. Gerber, J. Chem. Phys., 2007, 127, 094305.36 T. Iwahashi, T. Miyamae, K. Kanai, K. Seki, D. Lim and

Y. Ouchi, J. Phys. Chem. B, 2008, 112, 11936.37 P. Stange, K. Fumino and R. Ludwig, Angew. Chem., Int. Ed.,

2013, 52, 2990.

38 R. Hayes, S. Imberti, G. G. Warr and R. Atkin, Angew. Chem.,Int. Ed., 2012, 51, 7468.

39 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone,B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato,X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng,J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda,J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao,H. Nakai, T. Vreven, J. A. Montgomery Jr., J. E. Peralta,F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N.Kudin, V. N. Staroverov, R. Kobayashi, J. Normand,K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar,J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene,J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo,R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin,R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin,K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador,J. J. Dannenberg, S. Dapprich, A. D. Daniels,O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski andD. J. Fox, Gaussian 09, Revision A.1, Gaussian, Inc.,Wallingford, CT, 2009.

Paper PCCP

Publ

ishe

d on

30

May

201

3. D

ownl

oade

d by

Nat

iona

l Chi

ao T

ung

Uni

vers

ity o

n 28

/04/

2014

01:

57:0

8.

View Article Online